COMPARISON OF GLM, GLMM AND GEE POISSON MATHEMATICAL MODELING PERFORMANCE (Case Study: Number of Pulmonary Tuberculosis Patients in Indonesia in 2019-2021)

This study aims to compare the performance of data modeling with Poisson regression Generalized Linear Model (GLM), Mixed (GLMM), and Estimating Equation (GEE) modeling. The case used is a factor that affects number Pulmonary Tuberculosis cases in Indonesia panel data. Based on AIC criteria, smallest BIC RMSE GLMM models perform better than GLM GEE. In addition, also has coefficient determination value. results showed percentage population smoking Unmet Kesehatan had significant positive influence cases. contrast, households who access handwashing soap proper drinking water significantly negatively influenced Therefore, public needs awareness reduce cigarette consumption increase environmental access, such as adequate clean sources handwashing.